CIE LAB color space and K-means clustering algorithm for segmenting remote sensing images

Abstract EN

Remote sensing images contain several types of geological phenomena such as mountains, valleys, rivers, and faults, which require a considerable effort to accurately identify them from digital satellite imagery and use them in topographic mapping.

There exists big importance of the automatic segmentation methods of the remotely sensed color images.

Segmenting stage is considered as an initial stage in processing and analysis, topographic and land use mapping effectively and accurately, and any mistake at this stage has significant impacts on the rest of the subsequent stages of image processing such as extraction, classification and interpretation.-